Detection of Cancerous Lesion in Colposcopic Uterine Cervix Image

This paper works at segmentation of lesion observed in cervical cancer, which is the second most common cancer among women worldwide. The purpose of segmentation is to determine the location for a biopsy to be taken for diagnosis. Cervix cancer is a disease in which cancer cells are found in the tissues of the cervix. The acetowhite region is a major indicator of abnormality in the cervix image. This project addresses the problem of segmenting uterine cervix image into different regions. We analyze two algorithms namely Watershed and K-means clustering algorithm. These segmentations methods are carried over for the colposcopic uterine cervix image. The present result shows the clear segmentation output using K-means clustering algorithm. For illustration purpose, the results are demonstrated at the end of this paper.